ClawHub
Back to skill

Security audit

uno

Security checks across malware telemetry and agentic risk

Overview

This is a broad remote tool gateway that is disclosed, but it can use persistent credentials, invoke account-linked services, fetch untrusted skill content, and affect ratings without clear user confirmation boundaries.

Install only if you trust MCPMarket as a broker for broad tool calls and account authorization. Require confirmation before posts, purchases, deletions, financial or business changes, downstream OAuth linking, downloads, loading fetched skills, or submitting ratings. Remove ~/.uno/token when finished and separately revoke any linked downstream service access in MCPMarket or the connected service.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
87% confidence
Finding
The skill is presented primarily as a tool search/invocation interface, but it also exposes functionality to fetch full third-party skill content, file lists, ZIP download links, and repository URLs. That broader capability materially expands the trust boundary and can be abused to ingest untrusted prompt/code artifacts from external sources without the manifest clearly foregrounding that supply-chain risk.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill instructs the user to persist a bearer access token on disk in a fixed location, but does not clearly warn that the token is sensitive or discuss the risks of local credential theft, backups, shell history leakage, or multi-user system exposure. Even with restrictive file permissions, local disk storage increases the attack surface for token compromise and unauthorized API use.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.